Superposition in Optical Computing
|
|
- Samson Scott
- 6 years ago
- Views:
Transcription
1 Superposition in Optical Computing Superposition in Optical Computing B. Keith Jenkins Signal and Image Processing Institute MC-0272 University of Southern California, Los Angeles, California B. Keith Jenkins Signal and Image Processing Institute MC University of Southern California, Los Angeles, California and and C. Lee Giles Air Force Office of Scientific Research/NE Boiling AFB, D.C C. Lee Giles Air Force Office of Scientific Research/NE Bolling AFB, D.C ABSTRACT The design of an optical computer must be based on the characteristics of optics and optical tech- ABSTRACT The design of an optical computer must be based on the characteristics of optics and optical technology, and not on those of electronic technology. The property of optical superposition is considered and the implications it has in the design of computing systems is discussed. It can be exploited in the implementation of optical gates, interconnections, and shared memory. nology, and not on those of electronic technology. The property of optical superposition is considered and the implications it has in the design of computing systems is discussed. It can be exploited in the implementation of optical gates, interconnections, and shared memory. INTRODUCTION Fundamental differences in the properties of electrons and photons provide for expected differences in computational systems based on these elements. Some, such as the relative ease with which optics can implement regular, massively interconnections are well known. In this paper we examine how the property of superposition of optical signals in a linear medium can be exploited in building an optical or hybrid optical/electronic computer. This property enables many optical signals to pass through the same point in space at the same time without causing mutual interference or crosstalk. Since electrons do not have this property, this helps to shed more light on the role that optics could play in computing. We will separately consider the use of this property in interconnections, gates, and memory. INTRODUCTION Fundamental differences in the properties of electrons and photons provide for expected differences in computational systems based on these elements. Some, such as the relative ease with which optics can implement regular, massively interconnections are well known. In this paper we examine how the property of superposition of optical signals in a linear medium can be exploited in building an optical or hybrid optical/electronic computer. This property enables many optical signals to pass through the same point in space at the same time without causing mutual interference or crosstalk. Since electrons do not have this property, this helps to shed more light on the role that optics could play in computing. We will separately consider the use of this property in interconnections, gates, and memory. INTERCONNECTIONS A technique for implementing optical interconnections from one 2 -D array to another (or within the same array) has been described [Jenkins et al, 1984]. It utilizes two holograms in succession (Fig. 1). The holograms can be generated by a computer plotting device. The idea is to define a finite number, M, of distinct interconnection patterns, and then assemble the interconnecting network using only these M patterns. The second hologram of Fig. i consists of an array of facets, one for each of the M interconnection patterns. The first hologram contains one facet for each input node, and serves to address the appropriate patterns in the second hologram. INTERCONNECTIONS A technique for implementing optical interconnections from one 2-D array to another (or within the same array) has been described [Jenkins et a! ]. It utilizes two holograms in succession (Fig. 1). The holograms can be generated by a computer plotting device. The idea is to define a finite number, M, of distinct interconnection patterns, and then assemble the interconnecting network using only these M patterns. The second hologram of Fig. 1 consists of an array of facets, one for each of the M interconnection patterns. The first hologram contains one facet for each input node, and serves to address the appropriate patterns in the second hologram. It is the superposition property that makes this interesting. Note that many different signal beams can pass through the same facet of the second hologram at the same time without causing mutual interference. (All of these signals merely get shifted in the same direction and by the same amount.) This feature decreases the complexity of both holograms - The first because it only has to address M facets, the second hologram because it only has M facets. Let N be the number of nodes in the input and output arrays. The complexity (number of resolvable spots) of each hologram can be shown to be proportional to NM, with the proportionality constant being approximately 25 [Jenkins et al., 1984]. Using this as a model for interconnections in computing, a comparison can be made between the complexity of these optical interconnections with those of electronic VLSI for various It is the superposition property that makes this interesting. Note that many different signal beams can pass through the same facet of the second hologram at the same time without causing mutual interference. (All of these signals merely get shifted in the same direction and by the same amount.) This feature decreases the complexity of both holograms -- The first because it only has to address M facets, the second hologram because it only has M facets. Let N be the number of nodes in the input and output arrays. The complexity (number of resolvable spots) of each hologram can be shown to be proportional to NM, with the proportionality constant being approximately 25 [Jenkins et al., 1984]. Using this as a model for interconnections in computing, a comparison can be made between the complexity of these optical interconnections with those of electronic VLSI for various SPIE Vol 963 Optical Computing 88 (1988) / 407 SPIE Vol. 963 Optical Computing 88 (1988) / 407
2 interconnection networks. Results of this have been given in [Giles and Jenkins, 1986]. It is found that in general the optical interconnections have an equal or lower space complexity than electronic interconnections, with the difference becoming more pronounced as the connectivity increases. interconnection networks. Results of this have been given in [Giles and Jenkins, 1986]. It is found that in general the optical interconnections have an equal or lower space complexity than electronic interconnections, with the difference becoming more pronounced as the connectivity increases. 1 < IMAGING >-I I-- F. T. A INPUT INPUT - F.T FIRSTi_A/!, HOLOGRAM SECOND\ HOLOGRAM OUTPUT SECOND HOLOGRAM OUTPUT Fig. 1. Optical holographic system for interconnections. Fig. 1. Optical holographic system for interconnections. SHARED MEMORY SHARED MEMORY The same superposition principle can be applied to memory cells, where many optical beams can The same superposition principle can be applied to memory cells, where many optical beams can read the same memory location simultaneously. This concept could be useful in building a shared memory machine. For this concept, we first consider abstract models of computation based on shared memories. The reason for this approach is to abstract out inherent limitations of electronic technology (such as limited interconnection capability); in designing an architecture one would adapt the abstract model to the limitations of optical systems. These shared memory models are basically a ization of the Random Access Machine. The Random Access Machine (RAM) model [Aho, Hoperoft, and Ullman, 1974] is a model of sequential computation, similar to but less primitive than the Turing machine. The RAM model is a one -accumulator computer in which the instructions are not allowed to modify themselves. A RAM consists of a read -only input tape, a write -only output tape, a program and a memory. The time on the RAM is bounded above by a polynomial function of time on the TM. The program of a RAM is not stored in memory and is unmodifiable. The RAM instruction set is is small and consists of operations such as store, add, subtract, and jump if greater than zero; indirect addresses are permitted. A common RAM model is the uniform cost one, which assumes that each RAM instruction requires one unit of time and each register one unit of space. Shared memory models are based on global memories and are differentiated by their accessibility to memory. In Fig. 2 we see a typical shared memory model where individual processing elements (PE's) have variable simultaneous access to an individual memory cell. Each PE can access any cell of read the same memory location simultaneously. This concept could be useful in building a shared memory machine. For this concept, we first consider abstract models of computation based on shared memories. The reason for this approach is to abstract out inherent limitations of electronic technology (such as limited interconnection capability); in designing an architecture one would adapt the abstract model to the limitations of optical systems. These shared memory models are basically a iza- tion of the Random Access Machine. The Random Access Machine (RAM) model [Aho, Hopcroft, and Ullman, 1974] is a model of sequential computation, similar to but less primitive than the Turing machine. The RAM model is a one-accumulator computer in which the instructions are not allowed to modify themselves. A RAM consists of a read-only input tape, a write-only output tape, a program and a memory. The time on the RAM is bounded above by a polynomial function of time on the TM. The program of a RAM is not stored in memory and is unmodifiable. The RAM instruction set is is small and consists of operations such as store, add, subtract, and jump if greater than zero; indirect addresses are permitted. A common RAM model is the uniform cost one, which assumes that each RAM instruction requires one unit of time and each register one unit of space. Shared memory models are based on global memories and are differentiated by their accessibility to memory. In Fig. 2 we see a typical shared memory model where individual processing elements (PE's) have variable simultaneous access to an individual memory cell. Each PE can access any cell of 408 / SPIE Vol. 963 Optical Computing 88 (1988) 408 / SPIE Vol. 963 Optical Computing 88 (1988)
3 the global memory in unit time. In addition, many PE's can access many different cells of the global memory simultaneously. In the models we discuss, each PE is a slightly modified RAM without the input and output tapes, and with a modified instruction set to permit access to the global memory. A separate input for the machine is provided. A given processor can generally not access the local memory of other processors. the global memory in unit time. In addition, many PE's can access many different cells of the global memory simultaneously. In the models we discuss, each PE is a slightly modified RAM without the input and output tapes, and with a modified instruction set to permit access to the global memory. A separate input for the machine is provided. A given processor can generally not access the local memory of other processors. control (write) input control (write) input shared memory shared memorv memory f cell I memory cell inputs inputs 1 cell of an array outputs outputs Fig. 2. Conceptual diagram of shared Fig. 2. Conceptual diagram of shared memory models. memory models. Fig. 3. One memory cell of an array, showing multiple optical beams providing contention -free read access. Fig. 3. One memory cell of an array, showing multiple optical beams providing contention-free read access. The various shared memory models differ primarily in whether they allow simultaneous reads and/or writes to the same memory cell. The PRAC, random access computer [Lev, Pippenger and Valiant, 1981] does not allow simultaneous reading or writing to an individual memory cell. The PRAM, random access machine, [Fortune and Wyllie, 1978] permits simultaneous reads but not simultaneous writes to an individual memory cell. The WRAM, write random access machine, denotes a variety of models that permit simultaneous reads and certain writes, but differ in how the write conflicts are resolved. For example, a model by Shiloach and Vishkin (1981) allows a simultaneous write only if all processors are trying to write the same value. The paracomputer [Schwartz, 1980] has simultaneous writes but only "some" of all the information written to the cell is recorded. The models represent a hierarchy of time complexity given by The various shared memory models differ primarily in whether they allow simultaneous reads and/or writes to the same memory cell. The PRAC, random access computer [Lev, Pippenger and Valiant, 1981] does not allow simultaneous reading or writing to an individual memory cell. The PRAM, random access machine, [Fortune and Wyllie, 1978] permits simultaneous reads but not simultaneous writes to an individual memory cell. The WRAM, write random access machine, denotes a variety of models that permit simultaneous reads and certain writes, but differ in how the write conflicts are resolved. For example, a model by Shiloach and Vishkin (1981) allows a simultaneous write only if all processors are trying to write the same value. The paracomputer [Schwartz, 1980] has simultaneous writes but only "some" of all the information written to the cell is recorded. The models represent a hierarchy of time complexity given by TPRAC>TPRAM >WRMAAM where T is the minimum number of time steps required to execute an algorithm on each model. where T is the minimum number of time steps required to execute an algorithm on each model. More detailed comparisons are dependent on the algorithm [Borodin and Hoperoft, 1985]. More detailed comparisons are dependent on the algorithm [Borodin and Hopcroft, 1985]. SPIE Vol. 963 Optical Computing 88 (1988) / 409 SPIE Vol. 963 Optical Computing 88 (1988) / 409
4 In general, none of these shared memory are physically realizable because of actual fan-in limitations. As an an electronic example, the ultracomputer [Schwartz, 1980] is an architectural manifestation of the paracomputer that uses a hardwired Omega network between the PE's and memories; it simulates the paracomputer within a time penalty of O (Iog2n). logen ). The current IBM RP3 project is a continuation of the (initial) work on the ultracomputer. Optical systems could in principle be used to implement this memory read capability. As a simple example, a single 11-bit memory cell can be represented by one pixel of a 11-D or 22-D array; the bit could be represented by the state (opaque or transparent) of the memory cell. Many optical beams can simultaneously read the contents of this memory cell without contention (Fig. 3). In addi- tion to this an interconnection network is needed between the PE's and the memory, that can allow any PE to communicate with any memory cell, preferably in one step, and with no contention. A regular crossbar is not sufficient for this because fan-in to a given memory cell must be allowed. Figure 4 shows a conceptual block diagram of a system based on the PRAM model; here the memory array operates in reflection instead of transmission. The fan-in required of the interconnection network is also depicted in the figure. D PEs INTERCONNECTION MEMORY NETWORK (dynamic) Fig. 4. Block diagram of an optical architecture based on RAM models. INPUT ~ VECTOR OPTICS CROSSBAR CROSSBAR MASK OPTICS 7 i 7 -"" -^OUTPUT ÒUTPUT VECTOR VECTOR Fig. 5. Example of an optical crossbar interconnection network. Optical systems can potentially implement crossbars that also allow this fan-in. Several optical crossbar designs discussed in [Sawchuk, et al., 1986] exhibit fan-in capability. An example is the 410 // SPIE Vol. 963 Optical Computing 88 (1988)
5 optical crossbar shown schematically in Fig. 5; it is based on earlier work on optical matrix-vector multipliers. The 1-D array on the left could be optical sources (LED's or laser diodes) or just the location of optical signals entering from previous components. An optical system spreads the light from each input source into a vertical column that illuminates the crossbar mask. Following the crossbar mask, a set of optics collects the light transmitted by each row of the mask onto one element of the optical crossbar shown schematically in Fig. 5; it is based on earlier work on optical matrix -vector multipliers. The 1 -D array on the left could be optical sources (LED's or laser diodes) or just the location of optical signals entering from previous components. An optical system spreads the light from each input source into a vertical column that illuminates the crossbar mask. Following the crossbar mask, a set of optics collects the light transmitted by each row of the mask onto one element of the output array. The states of the pixels in the crossbar mask (transparent or opaque) determine the state of the crossbar switch. Multiple transparent pixels in a column provide fanout; multiple transparent pixels in a row provide fan-in. Many optical reconfigurable network designs are possible, and provide tradeoffs in performance parameters such as bandwidth, reconfiguration time, maximum number of lines, hardware requirements, etc. Unfortunately, most simple optical crossbars will be limited in size to approximately 256 x 256 (Sawchuk, et al., 1986). We are currently considering variants of this technique to increase the number of elements. Possibilities include using a multistage but nonblocking interconnection network (e.g. Clos), a hierarchy of crossbars, and /or a memory hierarchy. output array. The states of the pixels in the crossbar mask (transparent or opaque) determine the state of the crossbar switch. Multiple transparent pixels in a column provide fanout; multiple transparent pixels in a row provide fan-in. Many optical reconfigurable network designs are possible, and provide tradeoffs in performance parameters such as bandwidth, reconfiguration time, maximum number of lines, hardware requirements, etc. Unfortunately, most simple optical crossbars will be limited in size to approximately 256 x 256 (Sawchuk, et al., 1986). We are currently considering variants of this technique to increase the number of elements. Possibilities include using a multistage but nonblocking interconnection network (e.g. Clos), a hierarchy of crossbars, and/or a memory hierarchy. GATES Since the superposition property of optics only applies in linear media, it cannot in general be GATES Since the superposition property of optics only applies in linear media, it cannot in general be used for gates, which of course are inherently nonlinear. However, for important special cases superposition can allow many optical gates to be replaced with one optical switch. Consider again the situation depicted in Fig. 3, with the aperture being used as a switch or relay. The control beam opens or closes the relay; when the relay is closed (i.e., aperture is transparent), many optical signal beams can independently pass through the relay. If b represents the control beam and ai the signal beams, this in effect computes b a or b ai, depending on which state of b closes the relay, where denotes the Ar'm operation (Fig. 6). used for gates, which of course are inherently nonlinear. However, for important special cases superposition can allow many optical gates to be replaced with one optical switch. Consider again the situation depicted in Fig. 3, with the aperture being used as a switch or relay. The control beam opens or closes the relay; when the relay is closed (i.e., aperture is transparent), many optical signal beams can independently pass through the relay. If b represents the control beam and ai the signal beams, this in effect computes b -at- or b-a^, depending on which state of b closes the relay, where denotes the AND operation (Fig. 6). superimposed gate superimposed gate Fig. 6. One optical relay or superimposed gate versus individual gates Fig. 6. One optical relay or superimposed gate versus individual gates with a common input. with a common input. Using this concept, a set of gates with a common input in a single-instruction multiple-data (SIMD) machine can be replaced with one optical switch or "superimposed gate". An example of this is in the control signals; instead of broadcasting each instruction or control bit to all PR's, a fan-in from all PE's to a common control switch is performed. Thus, for / control bits per instruction word, / superimposed gates could replace NI gates (/ per PE). Since for optical or hybrid systems we Using this concept, a set of gates with a common input in a single- instruction multiple -data (SIMD) machine can be replaced with one optical switch or "superimposed gate ". An example of this is in the control signals; instead of broadcasting each instruction or control bit to all PE's, a fan-in from all PE's to a common control switch is performed. Thus, for I control bits per instruction word, I superimposed gates could replace NI gates (I per PE). Since for optical or hybrid systems we expect NCI, this can be a substantial reduction. Fig. 7 shows an example of how this can be incorporated into fixed optical interconnections (such as those of Fig. 1). In the figure there are four PE's laid out on a 2 -D array of gates. Each PE sends a signal through one pixel of a transmissive spatial light modulator (SLM). The SLM is electrically addressed, so that the instructions can come from an expect Af»/, this can be a substantial reduction. Fig. 7 shows an example of how this can be incor- porated into fixed optical interconnections (such as those of Fig. 1). In the figure there are four PE's laid out on a 2-D array of gates. Each PE sends a signal through one pixel of a transmissive spatial light modulator (SLM). The SLM is electrically addressed, so that the instructions can come from an SPIE Vol. 963 Optical Computing 88 (1988) / 411 SPIE Vol. 963 Optical Computing 88 (1988) / 411
6 SLMi gate array output SLM 1 gate array output hologram H1 hologram SLM2 H2 electrically controlled hologram SLM 2 electrically controlled gate input array gate input array Fig. 7. An optical architecture for the incorporation of superimposed gates for instruction or control Fig. 7. An optical architecture for the incorporation of superimposed gates for instruction or control bits. The optics are omitted for clarity but are identical to those of Fig. 1. Signals from four gates are shown that fan in to a common control bit. bits. The optics are omitted for clarity but are identical to those of Fig. 1. Signals from four gates are shown that fan in to a common control bit. electronic host. After passing through a common superimposed gate corresponding to the control bit, electronic host. After passing through a common superimposed gate corresponding to the control bit, the signals proceed to the appropriate gate inputs in the gate input array. In this case the second hologram H2 deflects the signals to the desired gate inputs (gates different from which they came). This optical system is identical to that of Fig. 1 except for the introduction of the SLM for control bits; thus the systems are compatable. Note also that the fanout of each gate in this process is one; a conventional implementation with a large number of PE's would require very high fanout capability or else a tree of gates for each control bit to provide the fanout. the signals proceed to the appropriate gate inputs in the gate input array. In this case the second hologram H 2 deflects the signals to the desired gate inputs (gates different from which they came). This optical system is identical to that of Fig. 1 except for the introduction of the SLM for control bits; thus the systems are compatable. Note also that the fanout of each gate in this process is one; a conven- tional implementation with a large number of PE's would require very high fanout capability or else a tree of gates for each control bit to provide the fanout. These superimposed gates are not true 3-terminal devices. The common (b) input is regenerated, but the at inputs are not. As a result, a design constraint, that these at signals do not go through too many superimposed gates in succession without being regenerated by a conventional gate, must be adhered to. This is typically not an issue in the case of control bits. Another consequence is that the total switching energy required for a given processing operation is reduced, because N gates are These superimposed gates are not true 3- terminal devices. The common (b ) input is regenerated, but the at inputs are not. As a result, a design constraint, that these ai signals do not go through too many superimposed gates in succession without being regenerated by a conventional gate, must be adhered to. This is typically not an issue in the case of control bits. Another consequence is that the total switching energy required for a given processing operation is reduced, because N gates are replaced with one superimposed gate. This is important because it is likely that the total switching energy will ultimately be the limiting factor on the switching speed and number of gates in an optical computer. Other advantages include an increase in computing speed since some of the gates are effectively passive and reduced requirements on the device used to implement the optical gates. replaced with one superimposed gate. This is important because it is likely that the total switching energy will ultimately be the limiting factor on the switching speed and number of gates in an optical computer. Other advantages include an increase in computing speed since some of the gates are effectively passive and reduced requirements on the device used to implement the optical gates. CONCLUSIONS We have shown that the property of superposition can be exploited in the design of optical or hybrid optical/electronic computing architectures. It can reduce the hologram complexity for highly interconnections, reduce the number of gates in a SIMD system, and permit simultaneous CONCLUSIONS We have shown that the property of superposition can be exploited in the design of optical or hybrid optical/electronic computing architectures. It can reduce the hologram complexity for highly interconnections, reduce the number of gates in a SIMD system, and permit simultaneous memory access in a shared memory machine, thereby reducing contention problems. Our fundamental reason for studying this is that architectures for optical computing must be designed for the capabilities and limitations of optics; they must not be constrained by the limitations of electronic memory access in a shared memory machine, thereby reducing contention problems. Our fun- damental reason for studying this is that architectures for optical computing must be designed for the capabilities and limitations of optics; they must not be constrained by the limitations of electronic 412 / SP /E Vol. 963 Optical Computing 88 (1988) 412 / SPIE Vol. 963 Optical Computing 88 (1988)
7 systems, which have necessarily dominated approaches to digital computing architectures to date. systems, which have necessarily dominated approaches to digital computing architectures to date. REFERENCES REFERENCES Aho, A.V., J.E. Hopcroft and J.D. Ullman, The Design and Analysis of Computer Algorithms, Reading, Mass. Addison-Wesley, Borodin, A. and J.E. Hopcroft, "Routing, Merging, and Sorting on Parallel Models of Computation." Journal of Computer and System Sciences, Vol. 30, pp Fortune, S., and J. Wyllie, "Parallelism in Random Access Machines," Proc 10th Annual ACM STOC, San Diego, California, pp , Giles, C. L., and Jenkins, B. K., "Complexity Implications of Optical Parallel Computing," Proc. Twentieth Annual Asilomar Conference on Signals, Systems, and Computers, Pacific Grove, Calif., Aho, A.V., J.E. Hoperoft and J.D. Ullman, The Design and Analysis of Computer Algorithms, Reading, Mass. Addison -Wesley, Borodin, A. and J.E. Hoperoft, "Routing, Merging, and Sorting on Parallel Models of Computation." Journal of Computer and System Sciences, Vol. 30, pp Fortune, S., and J. Wyllie, "Parallelism in Random Access Machines," Proc 10th Annual ACM STOC, San Diego, California, pp , Giles, C. L., and Jenkins, B. K., "Complexity Implications of Optical Parallel Computing," Proc. Twentieth Annual Asilomar Conference on Signals, Systems, and Computers, Pacific Grove, Calif., to appear, Nov Jenkins, B.K., et al., "Architectural Implications of a Digital Optical Processor," Applied Optics, Vol. 23, No. 19, pp , Lev, G., N. Pippenger, and L.G. Valiant, "A Fast Parallel Algorithm for Routing in Permutation Networks" IEEE Trans. on Computers, Vol. C -30, No. 2, pp , Feb Sawchuk, A. A., B.K. Jenkins, C.S. Raghavendra, and A. Vanua, "Optical Matrix -Vector Implementations of Crossbar Interconnection Networks," Proc. International Conference on Parallel Processing, St. Charles, IL, August 1986; also Sawchuk, et al., "Optical Interconnection Networks," Proc. International Conference on Parallel Processing, pp , August Schwartz, J.T., "Ultracomputers," A.C.M. Trans on Prog. Lang. and Sys., Vol. 2, No. 4, pp , October Shiloach, Y., and Vishkin, U., "Finding the Maximum, Merging and Sorting in a Parallel Computation Model," J. Algorithms, pp , March to appear, Nov Jenkins, B.K., et al., "Architectural Implications of a Digital Optical Processor," Applied Optics, Vol. 23, No. 19, pp , Lev, G., N. Pippenger, and L.G. Valiant, "A Fast Parallel Algorithm for Routing in Permutation Networks" IEEE Trans. on Computers, Vol. C-30, No. 2, pp , Feb Sawchuk, A. A., B.K. Jenkins, C.S. Raghavendra, and A. Varma, "Optical Matrix-Vector Implementations of Crossbar Interconnection Networks," Proc. International Conference on Parallel Processing, St. Charles, IL, August 1986; also Sawchuk, et al., "Optical Interconnection Networks," Proc. International Conference on Parallel Processing, pp , August Schwartz, J.T., "Ultracomputers," A.C.M. Trans on Prog. Lang. and Sys., Vol. 2, No. 4, pp , October Shiloach, Y., and Vishkin, U., "Finding the Maximum, Merging and Sorting in a Parallel Computation Model," /. Algorithms, pp , March SPIE Vol. 963 Optical Computing 88 (1988) / 413 SPIE Vol. 963 Optical Computing 88 (1988) / 413
A Sublinear Algorithm for Split-Merge Displacement Estimation on an Optical Architecture
A Sublinear Algorithm for Split-Merge Displacement Estimation on an Optical Architecture BRUNO CARPENTIERI Dipartimento di Informatica ed Applicazioni R. M. Capocelli Università di Salerno Via S. Alle
More informationThe p-sized partitioning algorithm for fast computation of factorials of numbers
J Supercomput (2006) 38:73 82 DOI 10.1007/s11227-006-7285-5 The p-sized partitioning algorithm for fast computation of factorials of numbers Ahmet Ugur Henry Thompson C Science + Business Media, LLC 2006
More informationRecognition. Clark F. Olson. Cornell University. work on separate feature sets can be performed in
Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 907-912, 1996. Connectionist Networks for Feature Indexing and Object Recognition Clark F. Olson Department of Computer
More informationEE/CSCI 451: Parallel and Distributed Computation
EE/CSCI 451: Parallel and Distributed Computation Lecture #4 1/24/2018 Xuehai Qian xuehai.qian@usc.edu http://alchem.usc.edu/portal/xuehaiq.html University of Southern California 1 Announcements PA #1
More informationSymbolic Manipulation of Boolean Functions Using a Graphical Representation. Abstract
Symbolic Manipulation of Boolean Functions Using a Graphical Representation Randal E. Bryant 1 Dept. of Computer Science Carnegie-Mellon University Abstract In this paper we describe a data structure for
More informationOptimization of Vertical and Horizontal Beamforming Kernels on the PowerPC G4 Processor with AltiVec Technology
Optimization of Vertical and Horizontal Beamforming Kernels on the PowerPC G4 Processor with AltiVec Technology EE382C: Embedded Software Systems Final Report David Brunke Young Cho Applied Research Laboratories:
More informationInvited Paper. Nukui-Kitamachi, Koganei, Tokyo, , Japan ABSTRACT 1. INTRODUCTION
Invited Paper Wavefront printing technique with overlapping approach toward high definition holographic image reconstruction K. Wakunami* a, R. Oi a, T. Senoh a, H. Sasaki a, Y. Ichihashi a, K. Yamamoto
More informationA Simple Placement and Routing Algorithm for a Two-Dimensional Computational Origami Architecture
A Simple Placement and Routing Algorithm for a Two-Dimensional Computational Origami Architecture Robert S. French April 5, 1989 Abstract Computational origami is a parallel-processing concept in which
More informationA CELLULAR, LANGUAGE DIRECTED COMPUTER ARCHITECTURE. (Extended Abstract) Gyula A. Mag6. University of North Carolina at Chapel Hill
447 A CELLULAR, LANGUAGE DIRECTED COMPUTER ARCHITECTURE (Extended Abstract) Gyula A. Mag6 University of North Carolina at Chapel Hill Abstract If a VLSI computer architecture is to influence the field
More informationAn Optical Data -Flow Computer
An ptical Data -Flow Computer Ahmed Louri Department of Electrical and Computer Engineering The University of Arizona Tucson, Arizona 85721 Abstract For many applications, such as signal and image processing,
More informationManArray Processor Interconnection Network: An Introduction
ManArray Processor Interconnection Network: An Introduction Gerald G. Pechanek 1, Stamatis Vassiliadis 2, Nikos Pitsianis 1, 3 1 Billions of Operations Per Second, (BOPS) Inc., Chapel Hill, NC, USA gpechanek@bops.com
More informationBinary computer-generated holograms (CGHs) are becoming essential components in optical signal processing schemes. Apart from their use as
High space bandwidth product computer-generated holograms using volume holography Joseph Rosen Optical Signal Processing Branch, Rome Laboratory, Hanscom Airforce Base, MA 01731, E-Mail: ROSEN@PL9000.PLH.AF.MIL,
More informationReliability Analysis of Gamma Interconnection Network Systems
International Journal of Performability Engineering, Vol. 5, No. 5, October 9, pp. 485-49. RAMS Consultants Printed in India Reliability Analysis of Gamma Interconnection Network Systems 1. Introduction
More informationDesign and Analysis of Kogge-Stone and Han-Carlson Adders in 130nm CMOS Technology
Design and Analysis of Kogge-Stone and Han-Carlson Adders in 130nm CMOS Technology Senthil Ganesh R & R. Kalaimathi 1 Assistant Professor, Electronics and Communication Engineering, Info Institute of Engineering,
More informationHybrid Integration of a Semiconductor Optical Amplifier for High Throughput Optical Packet Switched Interconnection Networks
Hybrid Integration of a Semiconductor Optical Amplifier for High Throughput Optical Packet Switched Interconnection Networks Odile Liboiron-Ladouceur* and Keren Bergman Columbia University, 500 West 120
More informationLiterature Survey of nonblocking network topologies
Literature Survey of nonblocking network topologies S.UMARANI 1, S.PAVAI MADHESWARI 2, N.NAGARAJAN 3 Department of Computer Applications 1 Department of Computer Science and Engineering 2,3 Sakthi Mariamman
More informationA 256-Radix Crossbar Switch Using Mux-Matrix-Mux Folded-Clos Topology
http://dx.doi.org/10.5573/jsts.014.14.6.760 JOURNAL OF SEMICONDUCTOR TECHNOLOGY AND SCIENCE, VOL.14, NO.6, DECEMBER, 014 A 56-Radix Crossbar Switch Using Mux-Matrix-Mux Folded-Clos Topology Sung-Joon Lee
More informationCOMP Parallel Computing. PRAM (4) PRAM models and complexity
COMP 633 - Parallel Computing Lecture 5 September 4, 2018 PRAM models and complexity Reading for Thursday Memory hierarchy and cache-based systems Topics Comparison of PRAM models relative performance
More informationMultiprocessor Interconnection Networks
Multiprocessor Interconnection Networks Todd C. Mowry CS 740 November 19, 1998 Topics Network design space Contention Active messages Networks Design Options: Topology Routing Direct vs. Indirect Physical
More informationBasic Processing Unit: Some Fundamental Concepts, Execution of a. Complete Instruction, Multiple Bus Organization, Hard-wired Control,
UNIT - 7 Basic Processing Unit: Some Fundamental Concepts, Execution of a Complete Instruction, Multiple Bus Organization, Hard-wired Control, Microprogrammed Control Page 178 UNIT - 7 BASIC PROCESSING
More informationUnit 9 : Fundamentals of Parallel Processing
Unit 9 : Fundamentals of Parallel Processing Lesson 1 : Types of Parallel Processing 1.1. Learning Objectives On completion of this lesson you will be able to : classify different types of parallel processing
More informationIE 495 Lecture 3. Septermber 5, 2000
IE 495 Lecture 3 Septermber 5, 2000 Reading for this lecture Primary Miller and Boxer, Chapter 1 Aho, Hopcroft, and Ullman, Chapter 1 Secondary Parberry, Chapters 3 and 4 Cosnard and Trystram, Chapter
More informationA Preliminary Version of an Optical Data-flow Architecture *
A Preliminary Version of an Optical Data-flow Architecture * Ahmed Louri Department of Electrical and Computer Engineering The University of Arizona Tucson, Arizona 85721 Abstract For many applications,
More informationEfficient Bitonic Communication for the Parallel Data Exchange
, October 19-21, 2011, San Francisco, USA Efficient Bitonic Communication for the Parallel Data Exchange Ralf-Peter Mundani, Ernst Rank Abstract Efficient communication is still a severe problem in many
More informationALGORITHMIC DECIDABILITY OF COMPUTER PROGRAM-FUNCTIONS LANGUAGE PROPERTIES. Nikolay Kosovskiy
International Journal Information Theories and Applications, Vol. 20, Number 2, 2013 131 ALGORITHMIC DECIDABILITY OF COMPUTER PROGRAM-FUNCTIONS LANGUAGE PROPERTIES Nikolay Kosovskiy Abstract: A mathematical
More informationTwo pixel computer generated hologram using a zero twist nematic liquid crystal spatial light modulator
Two pixel computer generated hologram using a zero twist nematic liquid crystal spatial light modulator Philip M. Birch, Rupert Young, David Budgett, Chris Chatwin School of Engineering, University of
More informationIncreased Underwater Optical Imaging Performance Via Multiple Autonomous Underwater Vehicles
DISTRIBUTION STATEMENT A. Approved for public release; distribution is unlimited. Increased Underwater Optical Imaging Performance Via Multiple Autonomous Underwater Vehicles Jules S. Jaffe Scripps Institution
More informationMassively Parallel Computing on Silicon: SIMD Implementations. V.M.. Brea Univ. of Santiago de Compostela Spain
Massively Parallel Computing on Silicon: SIMD Implementations V.M.. Brea Univ. of Santiago de Compostela Spain GOAL Give an overview on the state-of of-the- art of Digital on-chip CMOS SIMD Solutions,
More informationA SUPER-RESOLUTION MICROSCOPY WITH STANDING EVANESCENT LIGHT AND IMAGE RECONSTRUCTION METHOD
A SUPER-RESOLUTION MICROSCOPY WITH STANDING EVANESCENT LIGHT AND IMAGE RECONSTRUCTION METHOD Hiroaki Nishioka, Satoru Takahashi Kiyoshi Takamasu Department of Precision Engineering, The University of Tokyo,
More informationTopological Design of Minimum Cost Survivable Computer Communication Networks: Bipartite Graph Method
Topological Design of Minimum Cost Survivable Computer Communication Networks: Bipartite Graph Method Kamalesh V.N Research Scholar, Department of Computer Science and Engineering, Sathyabama University,
More informationDesign of Low Power Digital CMOS Comparator
Design of Low Power Digital CMOS Comparator 1 A. Ramesh, 2 A.N.P.S Gupta, 3 D.Raghava Reddy 1 Student of LSI&ES, 2 Assistant Professor, 3 Associate Professor E.C.E Department, Narasaraopeta Institute of
More informationArchitectures of Flynn s taxonomy -- A Comparison of Methods
Architectures of Flynn s taxonomy -- A Comparison of Methods Neha K. Shinde Student, Department of Electronic Engineering, J D College of Engineering and Management, RTM Nagpur University, Maharashtra,
More informationSupplementary Materials for
www.sciencemag.org/cgi/content/full/338/6107/640/dc1 Supplementary Materials for Quantum Entanglement of High Angular Momenta Robert Fickler,* Radek Lapkiewicz, William N. Plick, Mario Krenn, Christoph
More informationLow-Power FIR Digital Filters Using Residue Arithmetic
Low-Power FIR Digital Filters Using Residue Arithmetic William L. Freking and Keshab K. Parhi Department of Electrical and Computer Engineering University of Minnesota 200 Union St. S.E. Minneapolis, MN
More informationAC : HOT SPOT MINIMIZATION OF NOC USING ANT-NET DYNAMIC ROUTING ALGORITHM
AC 2008-227: HOT SPOT MINIMIZATION OF NOC USING ANT-NET DYNAMIC ROUTING ALGORITHM Alireza Rahrooh, University of Central Florida ALIREZA RAHROOH Alireza Rahrooh is a Professor of Electrical Engineering
More informationEfficient Self-Reconfigurable Implementations Using On-Chip Memory
10th International Conference on Field Programmable Logic and Applications, August 2000. Efficient Self-Reconfigurable Implementations Using On-Chip Memory Sameer Wadhwa and Andreas Dandalis University
More informationPerformance Study of Packet Switching Multistage Interconnection Networks
ETRI Journal, volume 16, number 3, October 1994 27 Performance Study of Packet Switching Multistage Interconnection Networks Jungsun Kim CONTENTS I. INTRODUCTION II. THE MODEL AND THE ENVIRONMENT III.
More informationDesigning Efficient Benes and Banyan Based Input-Buffered ATM Switches
Designing Efficient Benes and Banyan Based Input-Buffered ATM Switches Rajendra V. Boppana Computer Science Division The Univ. of Texas at San Antonio San Antonio, TX 829- boppana@cs.utsa.edu C. S. Raghavendra
More informationIntroduction to ATM Technology
Introduction to ATM Technology ATM Switch Design Switching network (N x N) Switching network (N x N) SP CP SP CP Presentation Outline Generic Switch Architecture Specific examples Shared Buffer Switch
More informationChapter 7. A modified concept of tri-state logic operation suitable for data handling in parallel computation
Chapter 7. A modified concept of tri-state logic operation suitable for data handling in parallel computation Abstract: A new concept of developing proper truth tables in ternary logic system is reported
More informationZero Order Correction of Shift-multiplexed Computer Generated Fourier Holograms Recorded in Incoherent Projection Scheme
VII International Conference on Photonics and Information Optics Volume 2018 Conference Paper Zero Order Correction of Shift-multiplexed Computer Generated Fourier Holograms Recorded in Incoherent Projection
More informationThe Complexity of FFT and Related Butterfly Algorithms on Meshes and Hypermeshes
The Complexity of FFT and Related Butterfly Algorithms on Meshes and Hypermeshes T.H. Szymanski McGill University, Canada Abstract Parallel FFT data-flow graphs based on a Butterfly graph followed by a
More informationFILTER SYNTHESIS USING FINE-GRAIN DATA-FLOW GRAPHS. Waqas Akram, Cirrus Logic Inc., Austin, Texas
FILTER SYNTHESIS USING FINE-GRAIN DATA-FLOW GRAPHS Waqas Akram, Cirrus Logic Inc., Austin, Texas Abstract: This project is concerned with finding ways to synthesize hardware-efficient digital filters given
More informationH&A Engineering. Systems
Introduction to Structured Methods The idea that system descriptions are more clear and easier with pictures rather than words provided the basis for the development of structured methods. Structured analysis
More informationDesign of Optical Burst Switches based on Dual Shuffle-exchange Network and Deflection Routing
Design of Optical Burst Switches based on Dual Shuffle-exchange Network and Deflection Routing Man-Ting Choy Department of Information Engineering, The Chinese University of Hong Kong mtchoy1@ie.cuhk.edu.hk
More informationChapter 2 Parallel Computer Models & Classification Thoai Nam
Chapter 2 Parallel Computer Models & Classification Thoai Nam Faculty of Computer Science and Engineering HCMC University of Technology Chapter 2: Parallel Computer Models & Classification Abstract Machine
More informationLED holographic imaging by spatial-domain diffraction computation of. textured models
LED holographic imaging by spatial-domain diffraction computation of textured models Ding-Chen Chen, Xiao-Ning Pang, Yi-Cong Ding, Yi-Gui Chen, and Jian-Wen Dong* School of Physics and Engineering, and
More informationPerformance of Multihop Communications Using Logical Topologies on Optical Torus Networks
Performance of Multihop Communications Using Logical Topologies on Optical Torus Networks X. Yuan, R. Melhem and R. Gupta Department of Computer Science University of Pittsburgh Pittsburgh, PA 156 fxyuan,
More informationSingle Pass Connected Components Analysis
D. G. Bailey, C. T. Johnston, Single Pass Connected Components Analysis, Proceedings of Image and Vision Computing New Zealand 007, pp. 8 87, Hamilton, New Zealand, December 007. Single Pass Connected
More informationVicinities for Spatial Data Processing: a Statistical Approach to Algorithm Design
Vicinities for Spatial Data Processing: a Statistical Approach to Algorithm Design Peter Y. Wu wu@rmu.edu Department of Computer and Information Systems Sushil Acharya acharya@rmu.edu Department of Engineering
More informationDesigning and Characterization of koggestone, Sparse Kogge stone, Spanning tree and Brentkung Adders
Vol. 3, Issue. 4, July-august. 2013 pp-2266-2270 ISSN: 2249-6645 Designing and Characterization of koggestone, Sparse Kogge stone, Spanning tree and Brentkung Adders V.Krishna Kumari (1), Y.Sri Chakrapani
More informationA quasi-nonblocking self-routing network which routes packets in log 2 N time.
A quasi-nonblocking self-routing network which routes packets in log 2 N time. Giuseppe A. De Biase Claudia Ferrone Annalisa Massini Dipartimento di Scienze dell Informazione, Università di Roma la Sapienza
More informationEE/CSCI 451: Parallel and Distributed Computation
EE/CSCI 451: Parallel and Distributed Computation Lecture #11 2/21/2017 Xuehai Qian Xuehai.qian@usc.edu http://alchem.usc.edu/portal/xuehaiq.html University of Southern California 1 Outline Midterm 1:
More information252C: Camera Stabilization for the Masses
252C: Camera Stabilization for the Masses Sunny Chow. Department of Computer Science University of California, San Diego skchow@cs.ucsd.edu 1 Introduction As the usage of handheld digital camcorders increase
More informationFourier, Fresnel and Image CGHs of three-dimensional objects observed from many different projections
Fourier, Fresnel and Image CGHs of three-dimensional objects observed from many different projections David Abookasis and Joseph Rosen Ben-Gurion University of the Negev Department of Electrical and Computer
More informationThe Effect of Adaptivity on the Performance of the OTIS-Hypercube under Different Traffic Patterns
The Effect of Adaptivity on the Performance of the OTIS-Hypercube under Different Traffic Patterns H. H. Najaf-abadi 1, H. Sarbazi-Azad 2,1 1 School of Computer Science, IPM, Tehran, Iran. 2 Computer Engineering
More informationProject - RTL Model. III. Design Decisions. Part 4 - May 17, Jason Nemeth. model is capable of modeling both provided benchmarks.
Project - RTL Model Part 4 - May 17, 2005 Jason Nemeth I. Introduction The purpose of this assignment was to refine the structural model of the microprocessor into a working RTL-level VHDL implementation.
More informationMeasurement of the Diffusion Coefficient of Miscible Fluids Using Both Interferometry and Wiener's Method 1
International Journal of Thermophysics, Vol. 22, No. 2, 2001 Measurement of the Diffusion Coefficient of Miscible Fluids Using Both Interferometry and Wiener's Method 1 N. Rashidnia, 2 R. Balasubramaniam,
More informationDT IC AD-A Sing H. Lee F.B *
AD-A260 051 Quarterly Progress Report for Design and Packaging of Fault Tolerant Optoelectronic Multiprocessor Computing System Sponsored by Defense Advanced Research Projects Agency Monitored by ONR Under
More informationDemonstration of the centralized optical backplane architecture in a three-board microprocessor-to-memory interconnect system
Available online at www.sciencedirect.com Optics & Laser Technology 3 (3) 127 131 www.elsevier.com/locate/optlastec Demonstration of the centralized optical backplane architecture in a three-board microprocessor-to-memory
More informationA Hybrid Interconnection Network for Integrated Communication Services
A Hybrid Interconnection Network for Integrated Communication Services Yi-long Chen Northern Telecom, Inc. Richardson, TX 7583 kchen@nortel.com Jyh-Charn Liu Department of Computer Science, Texas A&M Univ.
More informationChapter 7. Widely Tunable Monolithic Laser Diodes
Chapter 7 Widely Tunable Monolithic Laser Diodes We have seen in Chapters 4 and 5 that the continuous tuning range λ is limited by λ/λ n/n g, where n is the index change and n g the group index of the
More informationImplementation Of Quadratic Rotation Decomposition Based Recursive Least Squares Algorithm
157 Implementation Of Quadratic Rotation Decomposition Based Recursive Least Squares Algorithm Manpreet Singh 1, Sandeep Singh Gill 2 1 University College of Engineering, Punjabi University, Patiala-India
More informationPerformance of general graph isomorphism algorithms Sara Voss Coe College, Cedar Rapids, IA
Performance of general graph isomorphism algorithms Sara Voss Coe College, Cedar Rapids, IA I. Problem and Motivation Graphs are commonly used to provide structural and/or relational descriptions. A node
More informationADAPTIVE GRIDS FOR GEOMETRIC OPERATIONS
ADAPTIVE GRIDS FOR GEOMETRIC OPERATIONS fm. RANDOLPH FRANKLIN, Electrical, Computer and Systems Engineering Department, Rensselaer Polytechnic Institute, Troy, New York, USA S INTRODUCTION PATIAL AND TOPOLOGICAL
More informationPerformance Recovery in Direct - Mapped Faulty Caches via the Use of a Very Small Fully Associative Spare Cache
Performance Recovery in Direct - Mapped Faulty Caches via the Use of a Very Small Fully Associative Spare Cache H. T. Vergos & D. Nikolos Computer Technology Institute, Kolokotroni 3, Patras, Greece &
More informationOptoelectronic Multi-Chip Modules Based on Imaging Fiber Bundle Structures
Optoelectronic Multi-Chip Modules Based on Imaging Fiber Bundle Structures Donald M. Chiarulli a, Steven P. Levitan b, Matt Robinson c Departments of a Computer Science and b Electrical Engineering, University
More informationOPTIMIZATION OF FIR FILTER USING MULTIPLE CONSTANT MULTIPLICATION
OPTIMIZATION OF FIR FILTER USING MULTIPLE CONSTANT MULTIPLICATION 1 S.Ateeb Ahmed, 2 Mr.S.Yuvaraj 1 Student, Department of Electronics and Communication/ VLSI Design SRM University, Chennai, India 2 Assistant
More informationParallel Random-Access Machines
Parallel Random-Access Machines Marc Moreno Maza University of Western Ontario, London, Ontario (Canada) CS3101 (Moreno Maza) Parallel Random-Access Machines CS3101 1 / 69 Plan 1 The PRAM Model 2 Performance
More informationYossi Shiioach. STAN-CS January COMPUTER SCIENCE DEPARTMENT School of Humanities and Sciences STANFORD UNIVERSITY
UNION-MEMBER ALGORITHMS FOR NON-DISJOINT SETS by Yossi Shiioach STAN-CS-79-728 January 19 7 9 COMPUTER SCIENCE DEPARTMENT School of Humanities and Sciences STANFORD UNIVERSITY Union-Member Algorithms for
More informationMULTICAST or one-to-many communication is highly
IEEE TRANSACTIONS ON COMMUNICATIONS, VOL. 46, NO. 7, JULY 1998 957 On Blocking Probability of Multicast Networks Yuanyuan Yang, Member, IEEE, and Jianchao Wang Abstract Multicast is a vital operation in
More informationEfficient Majority Logic Fault Detector/Corrector Using Euclidean Geometry Low Density Parity Check (EG-LDPC) Codes
Efficient Majority Logic Fault Detector/Corrector Using Euclidean Geometry Low Density Parity Check (EG-LDPC) Codes 1 U.Rahila Begum, 2 V. Padmajothi 1 PG Student, 2 Assistant Professor 1 Department Of
More informationReductions of the general virus detection problem
EICAR 2001 Best Paper Proceedings Leitold, F. (2001). Reductions of the general virus detection problem. In U. E. Gattiker (Ed.), Conference Proceedings EICAR International Conference, (pp. 24-30). ISBN:
More informationVirtual Circuit Blocking Probabilities in an ATM Banyan Network with b b Switching Elements
Proceedings of the Applied Telecommunication Symposium (part of Advanced Simulation Technologies Conference) Seattle, Washington, USA, April 22 26, 21 Virtual Circuit Blocking Probabilities in an ATM Banyan
More informationParallel Random Access Memory in a Shared Memory Architecture
Parallel Random Access Memory in a Shared Memory Architecture Tran Duc Linh, Tony de Souza-Daw 1Centre of Technology RMIT University, Ho Chi Minh, Vietnam anthony.desouza-daw@rmit.edu.vn Thang Manh Hoang,
More informationDesign of paraxial diffractive elements with the CAD system DIGIOPT
Design of paraxial diffractive elements with the CAD system DIGIOPT Harald Aagedal, Thomas Beth, Heiko Schwarzer, Stephan Teiwes Institute of Algorithms and Cognitive Systems, University of Karlsruhe Am
More informationMULTICAST CONNECTION CAPACITY OF WDM SWITCHING NETWORKS WITHOUT WAVELENGTH CONVERSION
MULTICAST CONNECTION CAPACITY OF WDM SWITCHING NETWORKS WITHOUT WAVELENGTH CONVERSION B. CHIDHAMBARARAJAN a,1 K.KALAMANI a,2 N. NAGARAJAN b,2 S.K.SRIVATSA a,3 Department of Electronics and Communication
More informationTechniques of Noninvasive Optical Tomographic Imaging
Techniques of Noninvasive Optical Tomographic Imaging Joseph Rosen*, David Abookasis and Mark Gokhler Ben-Gurion University of the Negev Department of Electrical and Computer Engineering P. O. Box 653,
More informationGRIP: A parallel graph reduction machine
GRIP: A parallel graph reduction machine Simon L. Peyton-Jones, Chris Clack and Jon Salkild University College, London Abstract GRIP - Graph Reduction In Parallel - is a Fifth Generation machine designed
More informationMinimizing Power Dissipation during. University of Southern California Los Angeles CA August 28 th, 2007
Minimizing Power Dissipation during Write Operation to Register Files Kimish Patel, Wonbok Lee, Massoud Pedram University of Southern California Los Angeles CA August 28 th, 2007 Introduction Outline Conditional
More informationDUE to the high computational complexity and real-time
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, VOL. 15, NO. 3, MARCH 2005 445 A Memory-Efficient Realization of Cyclic Convolution and Its Application to Discrete Cosine Transform Hun-Chen
More informationFIRM: A Class of Distributed Scheduling Algorithms for High-speed ATM Switches with Multiple Input Queues
FIRM: A Class of Distributed Scheduling Algorithms for High-speed ATM Switches with Multiple Input Queues D.N. Serpanos and P.I. Antoniadis Department of Computer Science University of Crete Knossos Avenue
More informationA Parallel Algorithm for Minimum Cost Path Computation on Polymorphic Processor Array
A Parallel Algorithm for Minimum Cost Path Computation on Polymorphic Processor Array P. Baglietto, M. Maresca and M. Migliardi DIST - University of Genoa via Opera Pia 13-16145 Genova, Italy email baglietto@dist.unige.it
More informationRobust Reversible Multiplexer Design using. Mach-Zehnder Interferometer
Robust Reversible Multiplexer Design using Mach-Zehnder Interferometer M.Zaheer Ahmed #1, S.Md.Imran Ali #2, P.Nagashyam #3,T.S.D.V.Subbarao #4 # Assistant Professor, # Assitant Professor, # UG student,
More informationA Reconfigurable Multifunction Computing Cache Architecture
IEEE TRANSACTIONS ON VERY LARGE SCALE INTEGRATION (VLSI) SYSTEMS, VOL. 9, NO. 4, AUGUST 2001 509 A Reconfigurable Multifunction Computing Cache Architecture Huesung Kim, Student Member, IEEE, Arun K. Somani,
More informationDAVID M. KOPPELMAN 2735 East 65th Street Brooklyn, NY 11234
A SELF ROUTING PERMUTATION NETWORK DAVID M. KOPPELMAN 735 East 65th Street Brooklyn, NY 34 A. YAVUZ ORUÇ Electrical Engineering Department University of Maryland and Institute for Advanced Computer Studies
More informationA Reconfigurable Network Architecture For Parallel Prefix Counting
A Reconfigurable Network Architecture For Parallel Prefix Counting R. Lin S. Olariu Department of Computer Science Department of Computer Science SUNY at Geneseo Old Dominion University Geneseo, NY 14454
More informationScaling and Power Spectra of Natural Images
Scaling and Power Spectra of Natural Images R. P. Millane, S. Alzaidi and W. H. Hsiao Department of Electrical and Computer Engineering University of Canterbury Private Bag 4800, Christchurch, New Zealand
More informationEnabling Node Repair in Any Erasure Code for Distributed Storage
Enabling Node Repair in Any Erasure Code for Distributed Storage K. V. Rashmi, Nihar B. Shah, and P. Vijay Kumar, Fellow, IEEE Abstract Erasure codes are an efficient means of storing data across a network
More informationCHAPTER TWO LITERATURE REVIEW
CHAPTER TWO LITERATURE REVIEW 2.1 Introduction. This chapter provides in detail about the multiple access technologies and the OCDMA system. It starts with a discussion on various existing multiple-access
More informationA Performance Comparison of Five Algorithms for Graph Isomorphism
A Performance Comparison of Five Algorithms for Graph Isomorphism P. Foggia, C.Sansone, M. Vento Dipartimento di Informatica e Sistemistica Via Claudio, 21 - I 80125 - Napoli, Italy {foggiapa, carlosan,
More informationDesign and Implementation of Multistage Interconnection Networks for SoC Networks
International Journal of Computer Science, Engineering and Information Technology (IJCSEIT), Vol.2, No.5, October 212 Design and Implementation of Multistage Interconnection Networks for SoC Networks Mahsa
More informationAn Efficient Hybrid Parallel Prefix Adders for Reverse Converters using QCA Technology
An Efficient Hybrid Parallel Prefix Adders for Reverse Converters using QCA Technology N. Chandini M.Tech student Scholar Dept.of ECE AITAM B. Chinna Rao Associate Professor Dept.of ECE AITAM A. Jaya Laxmi
More informationOptimal Parallel Randomized Renaming
Optimal Parallel Randomized Renaming Martin Farach S. Muthukrishnan September 11, 1995 Abstract We consider the Renaming Problem, a basic processing step in string algorithms, for which we give a simultaneously
More informationParallel Systems Course: Chapter VIII. Sorting Algorithms. Kumar Chapter 9. Jan Lemeire ETRO Dept. November Parallel Sorting
Parallel Systems Course: Chapter VIII Sorting Algorithms Kumar Chapter 9 Jan Lemeire ETRO Dept. November 2014 Overview 1. Parallel sort distributed memory 2. Parallel sort shared memory 3. Sorting Networks
More informationSwitzerland ABSTRACT. Proc. of SPIE Vol N-1
Two-dimensional refractive index profiling of optical fibers by modified refractive near-field technique A. El Sayed* a,b, Soenke Pilz b, Manuel Ryser a, Valerio Romano a,b a Institute of Applied Physics,
More informationCHAPTER 1 INTRODUCTION
1 CHAPTER 1 INTRODUCTION 1.1 Advance Encryption Standard (AES) Rijndael algorithm is symmetric block cipher that can process data blocks of 128 bits, using cipher keys with lengths of 128, 192, and 256
More informationWORD LEVEL FINITE FIELD MULTIPLIERS USING NORMAL BASIS
WORD LEVEL FINITE FIELD MULTIPLIERS USING NORMAL BASIS 1 B.SARGUNAM, 2 Dr.R.DHANASEKARAN 1 Assistant Professor, Department of ECE, Avinashilingam University, Coimbatore 2 Professor & Director-Research,
More informationDiffraction and Interference
Experiment #32 Diffraction and Interference Goals: Perform a quantitative investigation of two-slit interference Explore use of a photodiode to measure light intensity References 1. I. G. Main, Vibrations
More informationDigital correlation hologram implemented on optical correlator
Digital correlation hologram implemented on optical correlator David Abookasis and Joseph Rosen Ben-Gurion University of the Negev Department of Electrical and Computer Engineering P. O. Box 653, Beer-Sheva
More information